ElasticSpace: A computational framework for interactive form-finding of textile hybrid structures through evolving topology networks. (18th December 2017)

• Record Type: Journal Article
• Authors: Suzuki, Seiichi; Knippers, Jan
• Is Part Of: International journal of parallel, emergent and distributed systems. Volume 32(2017)Supplement 1
• Page Start: S4
• Page End: S14
• Publication Date: 2017-12-18
• Subjects: Form-finding -- evolving networks -- textile hybrid structures -- dynamic relaxation -- dynamic topology -- data structure; Parallel computers -- Periodicals; Electronic data processing -- Distributed processing -- Periodicals; Computer algorithms -- Periodicals; 004.35
• DOI: 10.1080/17445760.2017.1390101
• Languages: English
• ISSNs: 1744-5760

Abstract

Abstract: Textile Hybrid Structures are a novel type of structural system referring to the coupling of tensile form- and bending-active components into a stiffer construct. For form finding its static equilibrium shape, several computational frameworks built upon the Dynamic Relaxation method have been developed for the interactive exploration of material and geometric properties. However, efforts are still required when addressing dynamic alterations of topology without completely resetting the simulation. The main problem to face is the dynamic alteration of topological data without losing consistency of connectivity. In this paper, we present the development of a computational framework for form-finding textile hybrid structures which enables dynamic explorations of complex topological configurations during solver's execution. A so-called evolving network formulation used to model mutable assemblies of interconnected particles is presented as well as the numerical scheme adopted to find the equilibrium state of such structures. The implementation of the framework is further described through the development of ElasticSpace, an interactive form finding tool for textile hybrid structures built with Java. Abstract : Topology-driven form-finding with ElasticSpace.